List of Papers Browse by Subject Areas Author List
Abstract
Superficial white matter (SWM) U-fibers contain consider able structural connectivity in the human brain; however, related studies are not well-developed compared to the well-studied deep white matter (DWM). Conventionally, SWM U-fiber is obtained through DWM tracking, which is inaccurate on the cortical surface. The significant variability in the cortical folding patterns of the human brain renders a conventional template-based atlas unsuitable for accurately mapping U-fibers within the thin layer of SWM beneath the cortical surface. Recently, new surface-based tracking methods have been developed to reconstruct more complete and reliable U-fibers. To leverage surface-based U-fiber tracking methods, we propose to create a surface-based U-fiber dictionary using high-resolution diffusion MRI (dMRI) data from the Human Connectome Project (HCP). We first identify the major U-fiber bundles and then build a dictionary containing subjects with high groupwise consistency of major U-fiber bundles. Finally, we propose a shape-informed U-fiber atlasing method for robust SWM connectivity analysis. Through experiments, we demonstrate that our shape-informed atlasing method can obtain anatomically more accurate U-fiber representations than state of-the-art atlas. Additionally, our method is capable of restoring incomplete U-fibers in low-resolution dMRI, thus helping better characterize SWM connectivity in clinical studies such as the Alzheimer’s Disease Neuroimaging Initiative (ADNI).
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/0213_paper.pdf
SharedIt Link: pending
SpringerLink (DOI): pending
Supplementary Material: N/A
Link to the Code Repository
N/A
Link to the Dataset(s)
N/A
BibTex
@InProceedings{Li_Surfacebased_MICCAI2024,
author = { Li, Yuan and Nie, Xinyu and Zhang, Jianwei and Shi, Yonggang},
title = { { Surface-based and Shape-informed U-fiber Atlasing for Robust Superficial White Matter Connectivity Analysis } },
booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
year = {2024},
publisher = {Springer Nature Switzerland},
volume = {LNCS 15002},
month = {October},
page = {pending}
}
Reviews
Review #1
- Please describe the contribution of the paper
1)They identify the major U-fiber bundles and then build a dictionary containing subjects with high groupwise consistency of major U-fiber bundles.2)They propose a shape-informed U-fiber atlasing method for robust SWM connectivity analysis. Through experiments, we demonstrate that our shape-informed atlasing method can obtain anatomically more accurate U-fiber representations than state of-the-art atlas.
- Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
The method in this paper can effectively avoid the problems of mismatch such as deformation that occurs when the SWM atlas is distorted to a new individual. In addition, the author also proves that the shape information can better match the individual differences than the traditional volume based U-fiber atlasing method.
- Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
This work point out that this paper studies a U-fiber atlas method that provides robust SWM connectivity analysis, but only demonstrates the validity and reliability of the method in the central sulcus of the motor cortex, and does not show the effect of other SWM regions.
- Please rate the clarity and organization of this paper
Satisfactory
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The submission has provided an anonymized link to the source code, dataset, or any other dependencies.
- Do you have any additional comments regarding the paper’s reproducibility?
In the “Dictionary of major U-fiber bundles with groupwise consistency” section,How to determine the threshold for clustering among individuals determined?
- Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html
In order to demonstrate the reliability of the method for other SWM regions, only the central sulcus was compared with the latest atlas in the paper, and only the motor cortex was mentioned in the later reproducibility experiment.
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making
Weak Reject — could be rejected, dependent on rebuttal (3)
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
This work constructed a dictionary of U fiber tracts with some reliability, while also taking into account individual variability. In this paper, the U-fiber atlas method based on shape information provides an idea for analyzing SWM connections, and the reliability of the method is confirmed to a certain extent.
- Reviewer confidence
Very confident (4)
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
Weak Reject — could be rejected, dependent on rebuttal (3)
- [Post rebuttal] Please justify your decision
The author refers to Superficial White Matter U-fibers throughout the article, including in the title. U-fibers are distributed across the entire brain, but the article focuses only on the central motor region. Even when considering the rebuttal mentioning the temporal lobe, this does not encompass the entire scope of the Superficial White Matter. I still have concerns about the validity of this map for the whole brain.
Review #2
- Please describe the contribution of the paper
This paper introduces a novel approach to surface-based and shape-informed atlas construction for robust analysis of Superficial white matter (SWM). Leveraging surface-based U-fiber tractography, the authors establish a comprehensive dictionary comprising subjects from the HCP data with strong groupwise consistency in major U-fibers. The incorporation of shape-informed atlasing enhances the reliability of SWM connectivity analysis, particularly in clinical investigations utilizing low-resolution diffusion MRI data. Overall, the proposed method promises progress in understanding SWM connectivity and its implications for clinical research.
- Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
The paper introduces a novel approach for creating an atlas of U-fiber bundles in the superficial white matter (SWM) of the brain to address the challenges posed by the complex cortical folding patterns. By leveraging surface-based tracking methods and high-resolution diffusion MRI data from the HCP data, the created dictionary-based atlas has high groupwise consistency. Results shows that the proposed method can successfully detect SWM connectivity changes in low-resolution diffusion MRI data from the ADNI.
- Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
The method assumes a certain level of homogeneity in cortical folding patterns across individuals, which may not hold true for all populations, especially under the low-resolution clinical data. While the method claims to restore incomplete U-fibers in low-resolution data, the extent and accuracy of this restoration are not thoroughly validated. Detailed analysis of restored U-fibers’ anatomical accuracy and comparison with high-resolution data are lacking.
- Please rate the clarity and organization of this paper
Good
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The submission does not mention open access to source code or data but provides a clear and detailed description of the algorithm to ensure reproducibility.
- Do you have any additional comments regarding the paper’s reproducibility?
no
- Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html
- Why the weight of distance transform is negative, while the shape index is positive in shape similarity score? Please clarify.
- After pull-back the identified U-fiber bundles from dictionary subjects, does the original streamlines retained？if so, how to solve the problem of false-positives in original U-fibers.
- The study does not mention the latest paper “Supwma: Consistent and Efficient Tractography Parcellation of Superficial White Matter with Deep Learning”, and there is no comparative experiment with this method.
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making
Weak Accept — could be accepted, dependent on rebuttal (4)
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The method is innovative, adding cortical shape information to traditional fiber classification method, and has made some clinical progress. However, the effectiveness of the method itself has not been very effectively verified.
- Reviewer confidence
Confident but not absolutely certain (3)
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
N/A
- [Post rebuttal] Please justify your decision
N/A
Review #3
- Please describe the contribution of the paper
- novel spherical fiber representation and computation of an atlas for cortico-cortico U-Fiber in the superficial white matter
- development of a U-Fiber atlas of 77 major U-fiber tracts
- spherical representation & shape-based dictionary for atlasing/clustering (rather than traditional volumetric registration based)
- Comparison with prior work (Roman et al 2022, NeuroImage)
- Please list the main strengths of the paper; you should write about a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
- Overal U-Fiber atlases are underdeveloped and this is a significant advanced relevant for the analysis of U-fiber tracts
- dictionary of U-fibers with group-wise consistency employed in the clustering
- shape similarity score patch matching
- Use of 755 HCP datasets to generate dictionary based atlas of 77 U-fibers
- Results look significantly better than comparison methods (and there are only few existing methods for this application) and overall well performing
- Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
- While the overall framework is novel, most of components/methods employed are not novel, overall somewhat iterative novelty
- Unclear whether the generated U-Fiber atlas will be available publicly and if so how it will be disseminated
- Please rate the clarity and organization of this paper
Very Good
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The submission does not mention open access to source code or data but provides a clear and detailed description of the algorithm to ensure reproducibility.
- Do you have any additional comments regarding the paper’s reproducibility?
Methods are well explained, data is publicly available (HCP and ADNI)
- Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html
- provide information on how the U-fiber atlas alongside the atlasing tool will be made available to the community
- No major revisions necessary to the document
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making
Accept — should be accepted, independent of rebuttal (5)
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
- difficult problem with no good existing solution. This paper presents an overall novel method that provides a working solution for U-fiber tracking & atlasing.
- evaluation is more than sufficient for a MICCAI paper
- Reviewer confidence
Very confident (4)
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
Accept — should be accepted, independent of rebuttal (5)
- [Post rebuttal] Please justify your decision
Appropriate rebuttal, my prior overall score was already “Accept”
Author Feedback
First, we would like to thank all reviewers for your time and valuable comments on our work. Following the suggestions of reviewers, we respond below to reviewers’ comments.
R1 C1: Open access to the community. Response: We will publicly release our atlas, source codes, and related data through the Neuroimaging Tools & Resources Collaboratory (NITRC) website.
R3 C1: Homogeneity of cortical folding. Response: We want to clarify that the variability of cortical folding across subjects is actually what motivates us to design our shape-informed atlasing method. We select the reference for a new subject based on the geometric information of cortical folding patterns while the previous atlases use all streamlines in the common space.
R3 C2: Impact of low-resolution diffusion MRI data. Response: We would like to clarity that our method relies on the cortical surface to propagate our atlas to new subject. Even if the dMRI data might have lower resolution than HCP, the T1-weighted MRI images typically have high resolution(~1mm) for cortical surface reconstruction. Thus, the application of our shape-informed U-fiber atlas does not depend on the resolution of the dMRI data.
R3 C3: Why the weight of distance transform is negative, while the shape index is positive in shape similarity score? Response: As shown in section 2.2, the weights of distance transform and shape index are both negative. These weights are negative by design for convenience.
R3 C4: After pull-back the identified U-fiber bundles from dictionary subjects, does the original streamlines retained? Response: All streamlines of the pull-backed U-fibers are from dictionary subjects while the original streamlines are not kept. Hence, we don’t deal with the false-positive of the original U-fibers.
R3 C5: The study does not mention the latest paper “Supwma: Consistent and Efficient Tractography Parcellation of Superficial White Matter with Deep Learning”, and there is no comparative experiment with this method. Response: Our study focuses more on building an accurate Superficial White Matter (SWM) dictionary and utilizing this dictionary while combining with the individual geometric information in practice. The work from Supwma relies on the anatomically curated white matter tractography atlas from the cited work [13] for the parcellation of SWM bundles, but the atlas in [13] still uses a volume-based tracking method, in contrast to our surface-based tracking method. We will cite Supwma as a complementary work of SWM studies in our revision and perform comparisons with this method in our future work.
R5 C1: Only central sulus and motor cortex were used for comparison with latest atlas and later experiments. Response: As shown in our work, our atlas includes 77 U-fiber bundles that cover most of the cortex, and the same shape-informed method can be applied to all bundles. We demonstrate our method on the U-fibers between the motor and sensory cortex because these regions are highly representative in previous SWM studies. However, the advantages derived from our surface-based tracking vs volume-based tracking in previous atlases are general for other bundles. In our current research, we have successfully applied our method in various cortical regions including the inferior, middle and superior temporal regions in both HCP and ADNI data. These results will be shared through the public distribution of the atlas on NITRC and future journal paper submission.
R5 C2: How to determine the threshold for inter-subject clustering in section 2.1? Response: The threshold is automatically determined in proportion to mean centroid length and the number of disconnected components from the hierarchical clustering method in [18].
Meta-Review
Meta-review #1
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Accept
- Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’
The authors have addressed the reviewers’ most pressing concerns in their rebuttal. Notably, the validity of this map for the U-fiber overall brain still needs to be clarified since both the manuscript and the rebuttal mentioned only a portion of the brain.
- What is the rank of this paper among all your rebuttal papers? Use a number between 1/n (best paper in your stack) and n/n (worst paper in your stack of n papers). If this paper is among the bottom 30% of your stack, feel free to use NR (not ranked).
The authors have addressed the reviewers’ most pressing concerns in their rebuttal. Notably, the validity of this map for the U-fiber overall brain still needs to be clarified since both the manuscript and the rebuttal mentioned only a portion of the brain.
Meta-review #2
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Accept
- Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’
The reviewers raised questions regarding open source, method details such as how to decide the threshold, and the rationale for selecting U-fibers for validation. They also proposed some additional methods for comparison with this paper. The authors answered these questions clearly in their feedback and stated that they will incorporate the methods proposed by the reviewers into the new version of the paper. Therefore, I suggest acceptance.
- What is the rank of this paper among all your rebuttal papers? Use a number between 1/n (best paper in your stack) and n/n (worst paper in your stack of n papers). If this paper is among the bottom 30% of your stack, feel free to use NR (not ranked).
The reviewers raised questions regarding open source, method details such as how to decide the threshold, and the rationale for selecting U-fibers for validation. They also proposed some additional methods for comparison with this paper. The authors answered these questions clearly in their feedback and stated that they will incorporate the methods proposed by the reviewers into the new version of the paper. Therefore, I suggest acceptance.